Thursday, 22 July 2010

Kelly: On The Edge

A couple of comments on my recent Kelly post, with some good debating points.

First Escapee:

Kelly Staking is near useless in the real world. One of its major flaws (and it has a few) is its accuracy is dependant on the accuracy of the 'edge' calculation which is just a guesstimate most of the time.
As I wrote, "the trick is calculating your edge in the first place", and while this may at first sight not seem to be easy to do, that does not mean that there is a flaw in the Kelly criterion. For profitable bettors, calculating the edge shouldn't be as hard as it sounds. I would agree that most gamblers do not know the true probabilities of their selections, and as a result they lose in the long run, but mathematically minded bettors who keep records and study their markets do know fairly accurately when they have an edge, and roughly how big that edge is. If they don’t, then they shouldn’t be betting! Since the reality of sports betting is that we seldom know the precise probabilities, and since over-betting has more serious consequences than under-betting, this is why I suggest fractional Kelly because it makes sense to err on the side of caution.
The object of a staking plan is to manage the risk of ruin of a betting bank against seeking maximum returns for that bank.
No staking system beats Kelly for this.
In the real world, Most of the People seeking out staking plans are trying out a 'system' that they don't even know if it has an edge let alone what figure it might be.
I would hope that anyone trying out a system would evaluate it based on level stakes over a reasonable length of time. As we have already agreed, Kelly cannot be used when the edge is an unknown.
Kelly is over complicated tosh, if you’re brainy enough to realise the importance of a staking plan to try out a system then all you need to do is set aside a bank for it and work out how many consecutive losses you want to allow the system before you admit its failure. Divide the bank by this figure and you have your bet size. Re-assess and adjust if the bank drops to 50% or doubles, repeat until bust or rich.
The problem with simply dividing your bank by a fixed number is that you are almost certainly going to go bust whereas if you use Kelly, the probability of busting is theoretically zero.
One of the so called advantages of Kelly is that it adjusts the bet size according to the near fictional edge calculation. But in reality this is poppycock! If you're running a system with odds or an 'edge' that requires a stake that varies enough to make any difference then you are actually running 2 ( or more ) different systems under the same bank and using Kelly will cloud the view of how each system is performing.
Not sure I understand how you work out that one can be running two or more different systems - but I do take the point that the edge will often not vary hugely from one selection to another, but sometimes it may.

Anonymous had a comment also, and a couple of good points:
For me, the main problem with Kelly is "bankroll"

What is your "bankroll"?

People will instantly answer "it's the amount of money you have set aside for betting"

Yeah right. These people would give up if they lost what they had set aside? They wouldn't dip into savings?

They'd never bet again.

That's why Kelly isn't practical. Mathematically it is sound although Cassini is right in saying fractional Kelly is normally best.
He is quite correct about bankrolls. It’s just meaningless to read on the forum, or anywhere for that matter, that someone has “blown their bank”. As Anonymous says, do these people just quit gambling (as arguably would be the sensible thing for them to do) or do they just cobble another bankroll together? Most people would do the latter. Gamblers are optimists in general.

I am still playing with my initial Betfair deposit from 2004. I don’t like throwing money away, and if I had lost that, which I came very close to doing at one point with a drawdown of 78.27%, I would have given up. I played with trading options and futures contracts for a while as I believe I have mentioned before, but when it became clear (as it should have been before I even began) that I had no edge over the professionals, I closed the account. No amount of reading books on trading options or futures was going to give me an edge, and for many it's the same with sports betting.
Oh, not calculating your perceived edge in running shouldn't be too difficult. You shouldn't be having a bet without an edge and to know you have an edge you need to go at least some of the way towards calculating it.
Absolutely correct. If you are betting, then presumably you’ve calculated that you have an edge. From there it’s a small step to come up with an estimate of how sizeable that edge is.

And if Kelly is just too much effort, just use 2%. Chances are you won't be too far off with that number.

1 comment:

Ross Taylor said...

You seem to have done well in following rule number one - never lose money